The Call for Proposals for RustWeek 2026 is now open!
If you’d like to give a talk, please submit your proposal via https://sessionize.com/rustweek-2026/
The CFP closes Dec 31.
TLDR; The PSF has made the decision to put our community and our shared diversity, equity, and inclusion values ahead of seeking $1.5M in new revenue. Please read and share. https://pyfound.blogspot.com/2025/10/NSF-funding-statement.html
🧵
https://www.python.org/sponsors/application/
Hello #VarnishCache users:
Release 8.0 is here:
And it even comes with bonus project news!
Next release will be march 15 but with a little twist...
So to my office, and there made my monthly accounts, and find myself worth in money about 686l. 19s. 2½d., for which God be praised.
In June, we announced HarfRust, a fully safe port of #HarfBuzz to Rust. At that time, HarfRust was 2x to 4x slower than HarfBuzz for a variety of benchmarks, so we have been working on addressing that.
Today, Chad Brokaw and I are pleased to present HarfRust 0.2.0, which is less than 25% slower than HarfBuzz, on both OpenType and AAT shaping benchmarks. We have also addressed all known correctness issues.
Charts:
https://docs.google.com/spreadsheets/d/1lyPPZHXIF8gE0Tpx7_IscwhwaZa4KOpdt7vnV0jQT9o/preview
so i'm reverse engineering the ipod's photo database file, right (y'know how on the old ipods you could sync photos to them if they had a color screen, and then you could browse your albums on the ipod? the internal database it creates for that)
as i've been building a parser for it, i've been nominally parsing things i don't understand yet into u32 fields until i can reason otherwise
anyway, as i'm poring over the parsed data, i see some interesting neighboring values, exactly four u32s, which i don't quite understand and which aren't documented (at least not in the docs on the iPodLinux wiki; they're incomplete and very outdated but they've still been an invaluable starting point)
for some of the photos (and for all album art in the artworkdb, which uses the same database format) these four u32 fields have constant values (i.e. matching between photos) but for other photos they look almost entirely random. perhaps they're hashes or randomized identifiers? but even for the random-looking ones they're kinda similar between photos
and then i think.. are they four floats maybe?
i plug the raw hex values into https://float.exposed (excellent website, highly recommend) but i'm not entirely convinced by the numbers i see
hmm.
the ones with constant values repeat on an 8-byte period, so maybe it's a pair of doubles?
so i tweak my parser to try reading them as such, re-run it, and am immediately filled with a horrible sense of dread as i recognize exactly what the numbers are.
.
.
.
IT;S FUCKING GEOTAGS AGAIN
Attention #bornhack! We have just published a new version of BORNHACK ADVENTURE, the premier #hacker camp text adventure: https://adventure.funhack.dk/
shit fuck i accidentally installed garfana instead of grafana
Garfield's worst nightmare (or how a API difference caused a bad bug)
by Valdemar Erk at RustWeek 2025!
https://www.youtube.com/watch?v=6aQdedw3a-0
** Speaker announcement **
Catch this talk at RustWeek 2025!
Speaker: Valdemar Erk
Title: Garfields worst nightmare (or how a API difference caused a bad bug)
Info & tickets: https://rustweek.org/talks/valdemar/
See you in Utrecht 13-17 May, 2025!
Anne C.A. Baanen
describing trains like user agent strings
Stevenson's Rocket (compatible; standard gauge; 320 km/h) Class/374 (Siemens Velaro, like ICE 3M) Eurostar/e320
Boost if you want less generative AI in your tech in 2025.
Got libsoba, Inochi2D's new UI toolkit running on #FreeBSD :)